WO2022046336A1 - System and method for machine monitoring - Google Patents
System and method for machine monitoring Download PDFInfo
- Publication number
- WO2022046336A1 WO2022046336A1 PCT/US2021/043108 US2021043108W WO2022046336A1 WO 2022046336 A1 WO2022046336 A1 WO 2022046336A1 US 2021043108 W US2021043108 W US 2021043108W WO 2022046336 A1 WO2022046336 A1 WO 2022046336A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- machine
- events
- data
- types
- event
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
- G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
- G05B19/0428—Safety, monitoring
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- E—FIXED CONSTRUCTIONS
- E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
- E02F—DREDGING; SOIL-SHIFTING
- E02F9/00—Component parts of dredgers or soil-shifting machines, not restricted to one of the kinds covered by groups E02F3/00 - E02F7/00
- E02F9/26—Indicating devices
- E02F9/267—Diagnosing or detecting failure of vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L1/00—Measuring force or stress, in general
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P15/00—Measuring acceleration; Measuring deceleration; Measuring shock, i.e. sudden change of acceleration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
- G01P3/00—Measuring linear or angular speed; Measuring differences of linear or angular speeds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24015—Monitoring
Definitions
- the present disclosure relates generally to machine or work systems, and more particularly, to methods and systems for monitoring the operation of one or more machines.
- Mobile machines e.g., earthmoving equipment
- stationary machines e.g., power-generation devices
- Such machines whether mobile or stationary, incorporate robust systems that allow the machine to perform one or more desired tasks or work (e.g., suspension systems, combustion systems, implement systems, power storage systems, etc.).
- Mobile machines such as trucks (e.g., mining trucks, haul trucks, on-highway trucks, off-highway trucks, articulated trucks), excavators, wheel loaders, other earthmoving equipment, as well as other mobile machines, experience a wide variety of physical loads and changing environments. Machines may even experience damage based on the actions of an operator controlling the machine, especially when the machine is operated by relatively inexperienced operators.
- U.S. Patent No. 10,466,690 B2 to Unuma et al. (the ’690 patent).
- the ’690 patent describes a device that estimates damage by determining changes in the operating status of a machine, for example, when an engine is running or the machine is stopped. These status changes may be insufficient to reflect events involving an environment in which the machine operates or other aspects of the operation of the machine.
- the damage estimation device described in the ’690 patent does not employ machine learning, which may reduce the accuracy, number, and/or types of tasks it is able to evaluate. Therefore, while the damage estimation device described in the ’690 patent may be useful in some circumstances, it may be unable to accurately identify types of events that significantly affect the wear and/or damage accumulated in components of the machine.
- the disclosed method and system may solve one or more of the problems set forth above and/or other problems in the art.
- the scope of the current disclosure is defined by the attached claims, and not by the ability to solve any specific problem.
- a method for monitoring operation of a machine may include receiving machine operation data including a first data channel indicative of a sensed condition at a first location on the machine, wherein the machine operation data is collected during one or more work cycles of the machine, and extracting at least a portion of the machine operation data, including data associated with the first data channel.
- the method may also include classifying the extracted machine operation data to identify one or more types of events that occurred during the one or more work cycles of the machine and involving an environment in which the machine operates during the one or more work cycles of the machine, estimating a condition at a second location on the machine during the one or more work cycles of the machine based on the identified one or more types of events, and logging the identified one or more types of events with a plurality of additional identified types of events to represent a plurality of work cycles during operation of the machine over time.
- a system for monitoring operation of a machine may include at least one processor and at least one non-transitory computer readable medium storing instructions which, when executed by the at least one processor, cause the one or more processors to embody circuitry including an input circuitry configured to receive machine operation data including a first data channel indicative of a sensed condition at a first location on the machine, the machine operation data corresponding to one or more work cycles of the machine and a data extraction circuitry for extracting at least a portion of the machine operation data, including data associated with the first data channel.
- the instructions may further cause the one or more processors to embody circuitry including an event identification circuitry for identifying one or more types of events involving an environment in which the machine operates and that occurred during the one or more work cycles of the machine by classifying the extracted machine operation data, a condition estimation circuitry for estimating a condition at a second location on the machine during the one or more work cycles of the machine based on the identified one or more types of events, and an event logging circuitry for logging the identified one or more types of events with a plurality of additional identified types of events to represent a plurality of work cycles during operation of the machine over time.
- an event identification circuitry for identifying one or more types of events involving an environment in which the machine operates and that occurred during the one or more work cycles of the machine by classifying the extracted machine operation data
- a condition estimation circuitry for estimating a condition at a second location on the machine during the one or more work cycles of the machine based on the identified one or more types of events
- an event logging circuitry for logging the identified one or
- a method for monitoring an operation of a machine may include receiving machine operation data corresponding to one or more work cycles of the machine, identifying one or more events in the machine operation data that occurred during the one or more work cycles of the machine and involved an environment of the machine, and classifying the identified one or more events with a type of event via a trained machine learning model.
- the method may further include determining an amount of damage at a location of the machine during the one or more work cycles of the machine based on the type of event, wherein no sensor is present at the location, and monitoring the operation of the machine over time by storing the type of event and a plurality of additional types of events in one or more databases.
- FIG. l is a diagram illustrating a system environment for including a work cycle monitoring system according to aspects of the present disclosure.
- FIG. 2 is diagram of a system environment for event classification, event tracking, and/or damage tracking, according to aspects of the present disclosure.
- FIG. 3 is an illustration of damage estimation and/or remaining life tracking for one or more locations of a machine, according to aspects of the present disclosure.
- FIG. 4 is a table illustrating exemplary simulation event data, according to aspects of the present disclosure.
- FIG. 5A is a series of charts illustrating exemplary data channels useful for identifying and characterizing an event, according to aspects of the present disclosure.
- FIG. 5B is a chart illustrating time-series data useful for identifying one or more types of events and exemplary identified events.
- FIG. 6 is a chart illustrating validation simulations associated an exemplary machine learning model, according to aspects of the present disclosure.
- FIG. 7 is a chart illustrating performance of exemplary machine learning algorithms, according to aspects of the present disclosure.
- FIG. 8 is a table illustrating exemplary work data, according to aspects of the present disclosure.
- FIG. 9 is a heatmap illustrating exemplary operator coaching and/or maintenance data, according to aspects of the present disclosure.
- FIG. 10 is a flowchart illustrating an exemplary process, according to aspects of the present disclosure.
- FIG. 11 is a block diagram illustrating an implementation of a computer system that may execute techniques, according to aspects of the present disclosure.
- FIG. 1 is a diagram illustrating a system environment 100 including a work cycle monitoring system 130 that may be useful for monitoring work cycles and damage in one or more machines 112.
- System environment 100 may include one or more machine systems 110 and respective machines 112, a network 170, and work cycle monitoring system 130.
- System environment 100 may further include one or more supervision systems 180 and one or more machine displays or user displays 190.
- Machines 112 may be any mobile or stationary machine for performing tasks relevant to boring, earthmoving, hauling, material handling, forestry, power generation, or other applications.
- Exemplary machines 112 may include trucks, dozers, excavators, pavers, loaders, draglines, drills, graders, generators, or any other suitable machine.
- Monitoring system 130 may monitor machines 112 for damage that occurs to one or more machine systems (e.g., hydraulic systems, implement systems, drivetrain systems, frames, suspension systems, ground-traversing systems, power generation or storage systems, etc.), or components thereof.
- machine systems e.g., hydraulic systems,
- Each machine 112 may be part of a machine system 110 that also includes an on-board system 120 which may be secured to machine 112, and one or more sensors secured at sensor locations 114, exemplary first locations, of machine 112.
- Each machine 112 may include an internal combustion engine (not shown) that is controlled by one or more electronic control modules (ECMs).
- ECMs electronice control modules
- an on-board system 120 may be configured to monitor the operation of the machine 112, including an operation of the machine 112, including travel speeds, positions, inclinations, and/or accelerations, among others.
- on-board system 120 may be configured to monitor conditions of the internal combustion engine, hydraulic systems, implement systems, drivetrain systems, frames, and/or suspension systems, among others.
- sensor locations 114 may include strain sensors, accelerometers, pressure sensors (e.g., positioned to measure hydraulic pressure for controlling one or more implements or suspension components), etc.
- one or more location systems e.g., GPS-based systems, radar-based proximity awareness systems, or others
- Each of these sensors and/or systems may be monitored by on-board system 120.
- On-board system 120 may be implemented within an ECM of machine system 110 or instead as one or more stand-alone systems.
- On-board system 120 may include appropriate circuitry (e.g., signal processing circuitry for receiving, analyzing, and transmitting data according to signals generated by sensors at locations 114), and a network or communication interface for communicating with one or more external (e.g., off-board) systems via network 170.
- on-board system 120 may include network interface circuitry for communicating with work cycle monitoring system 130, supervision system 180, and machine or user displays 190.
- Work cycle monitoring system 130 may be configured to monitor a plurality of work cycles of at least one machine 112, and preferably a plurality of similar or identical machines 112, or, if desired, a plurality of different types of machines 112.
- Monitoring system 130 may include one or more components or systems configured to identify events that occur during the operation of machine 112 and classify or map these events to one or more predetermined types of events, as described in detail below, including appropriate circuitry for accomplishing each function described herein.
- monitoring system 130 may include a machine learning (“ML”) trainer 140 configured to create, modify, and use datasets for training a machine learning model 150.
- ML trainer 140 may include one or more trainer databases (“DB”) 142 storing training data, and a multi -body dynamics simulation 144 that represents a particular machine 112.
- Multi-body dynamics simulation 144 may include one or more physics-based simulation models configured to generate simulated transient data associated with predetermined events that may be experienced by machine 112, as described below. Simulated data generated by multi -body dynamics simulation 144 may be stored in trainer DB 142, for example.
- Monitoring system 130 may also include one or more ML DBs 152 configured to store information output from ML model 150.
- ML DBs 152 may include one or more datasets corresponding to events identified with ML model 150, and respective instantaneous and/or accumulated damages (e.g., based on one or more calculated amounts of strain) corresponding to locations of machine 112.
- each instantaneous and/or accumulated damage may correspond to a sensor location 114 of machine 112 (e.g., a location on which a strain sensor or other type of sensor is located), or to a sensor-free location 116.
- Sensor-free locations 116 may correspond to exemplary second locations where no sensor is located.
- An event extractor 154 may be configured to monitor machine operation data from sensors and an ECM associated with machine 112 and identify one or more events that occurred during operation of machine 112.
- a work cycle compiler 156 may be configured to monitor a status of machine 112 according to damage that accumulates over time according to estimates output by ML model 150. Work cycle compiler 156 may be configured to generate notifications associated with work cycle composites, machine maintenance, operator training, damage aggregation, and/or damage validation, as described below.
- Monitoring system 130 may be configured to communicate with one or more supervision systems 180 for monitoring machine 112 or a fleet (plurality) of machines 112.
- Supervision systems 180 may be configured to display a status (e.g., identified events, an accumulated damage, remaining life, need for maintenance or repair, etc.) to one or more users, such as fleet managers or maintenance personnel.
- monitoring system 130 may be configured to cause one or more machine or user displays 190 to display a status or notification.
- Machine or user displays 190 may correspond to a user device associated with a user (e.g., an operator of machine 112, maintenance personnel, etc.), including a mobile system (smart phone, laptop) or other computer system. Additionally or alternatively, display 190 may correspond to a display of machine 190.
- One or both of supervision system 180 and display 190 may display a suitable notification that provides information regarding one or more damaged components or systems of machine 112, a need to perform maintenance on machine 112, or any other suitable notifications.
- work cycle monitoring system 130 is illustrated as being an off-board (e.g., remote) system that communicates with on-board system 120, in at least some configurations of system environment 100, one or more components of monitoring system 130 (e.g., ML model 150, event extractor 154, and work cycle compiler 156), may be provided as on-board components of machine 112. These components of monitoring system 130 may, for example, be implemented by on-board system 120, or a separate on-board system. If desired, ML trainer 140 and/or ML DB 152 may also be implemented in machine 112 as on-board components. FIG.
- Machine data 210 may be an input received by monitoring system 130 via network 170 in a plurality of channels (four channels shown in FIG. 2), including a first data channel corresponding to a sensor location 114. This first data channel may be indicative of a condition sensed by a strain sensor or any other sensor described herein.
- the first data channel, or any other data channel of machine data 210 may be indicative of information received from the ECM (e.g., engine speed, fuel rate, operating temperatures, or others).
- Machine data 210 may be received in real-time or near real-time. However, if desired, machine data 210 may be received after a period of time, and/or may be received intermittently (e.g., daily, weekly, monthly, etc., depending on the availability of network 170, available bandwidth, and/or availability of processing capacity).
- One or more algorithms may be employed by monitoring system 130 to identify the occurrence of an event, and extract the corresponding machine data 210 before the type of event is identified.
- an events, features extractor 220 (“EF extractor” 220), which may correspond to event extractor 154, may identify the occurrence of events reflected in machine data 210 and may extract data 210 containing these events and one or more features (e.g., event signatures reflected in one data channels corresponding to a particular sensor and/or condition output from an ECM, such as machine speed, payload, strut pressures, etc.) for analysis by one or more ML models 150 (e.g., event classifier models 250 and regression models 240).
- ML models 150 e.g., event classifier models 250 and regression models 240.
- This extraction may be performed, for example, based on events simulated by multi-body dynamics simulation 234, by searching for event signatures in machine data 210 that match simulated event signatures, without performing an identification of the type of event reflected in data 210.
- event signatures may be present in sensed data, data received from an ECM, or both.
- Event signatures may correspond to, for example, sensed pressures (e.g., pressures on struts, shocks, springs, wheels, axles, body, frame), lateral accelerations and/or velocities, forward or reverse accelerations and/or velocities, etc., that exceed a predetermined threshold or match an expected waveform (e.g., a reverse acceleration that transitions to forward motion or a stop over a predetermined period of time).
- sensed pressures e.g., pressures on struts, shocks, springs, wheels, axles, body, frame
- lateral accelerations and/or velocities e.g., forward or reverse accelerations and/or velocities, etc.
- a database or matrix of predetermined events 232 may store data corresponding to various types of events that machine 112 may be expected to experience, and may be provided to multi-body dynamics simulation 234 for simulation of these events. These simulations may be useful for determining forces, wear, and/or damages at various locations of machine 112 during each event.
- Data contained in events matrix 232 may also be provided to an ML classification trainer 260, as shown in FIG. 2.
- event labels examples of which are described below and shown in FIG. 4, may be output to ML classification trainer 260 from events matrix 232.
- Multi-body dynamics simulation 234 may correspond to multibody dynamics simulation 144, described above with respect to FIG. 1, and may include a plurality of simulated structural bodies of machine 112 that, when subjected to simulated transient events, produce outputs in the form of dynamic structural responses. Events simulated in multi-body dynamics simulation 234 may be associated with one or more predetermined types of events specified by events matrix 232. Exemplary structural responses may include strains, reactionary forces, such as bearing loads on pins or bearings, stresses, etc., at multiple locations of the simulated machine 112. Multi-body dynamics simulation 234 may include structural responses of the machine through the use of flexible bodies, such as flexible bodies that represent one or more components of the frame of machine 112.
- multi-body dynamics simulation 234 may simulate a dump body, systems for raising and lowering the dump body, and suspension components of the machine 112, including members, hinges, linkages, and/or struts.
- the simulated data produced by multi-body dynamics simulation 234, in particular, damage that occurred at various locations during the simulated event, may be output to a simulated structural responses DB 236 for use with ML regression trainer 230.
- These simulated structural responses may also be provided as an output to EF extractor 220.
- EF extractor 220 may output detected or actual events to a corresponding database, referred to herein as actual events, features DB 222 (“actual EF DB” 222).
- Actual EF DB 222 may be used to store one or more datasets that includes each extracted field event (e.g., actual events contained in machine data 210 and extracted by EF extractor 220), as well as corresponding features associated with each channel of data.
- EF extractor 220 may be further configured to extract simulated events and features from simulations performed with multi-body dynamics simulation 234. These simulated events and features (e.g., simulated data channels) may be stored as one or more datasets in simulated events, features database 224 (“simulated EF DB” 224).
- One or more event classifier models 250 may be configured to classify one or more extracted events based on information received from actual events, features DB 222 and training performed with ML classification trainer 260. In at least some configurations, a plurality of event classifier models 250 may be employed, each of these models 250 being trained to classify a particular type of event. In the example of a mining operation, types of events may include dumping, loading, potholes, ditches, turning, and any other event described herein or suitable for monitoring one of the machines described herein.
- ML classification trainer 260 may be configured to receive simulated data from simulated EF DB 224 and labels from predetermined events matrix 232. Thus, ML classification trainer 260 may employ simulated data (e.g., data from simulated EF DB 224) for ground truth labels, and event signatures as features.
- Event classifier models 250 may be configured to output classified or identified events to events frequency DB 270, which may store datasets that represent a number of occurrences of each type of event and damage incurred during the event at a plurality of locations such as locations 114 and 116 (FIG. 1). This data may be output to event tracker 280 which may log a plurality of events in a composite work cycles database 282 to monitor how many times particular events occurred and how often these events occurred for a plurality of machines 112. Additionally, events frequency DB 270 may be configured to output identified types of events and, if desired, corresponding damages at one or more locations of machine 112, to a coaching notification generator 272 which may be configured to generate training notifications to one or both of supervision system 180 and machine/user display 190. Additionally or alternatively, events frequency DB 270 may include datasets provided to maintenance notification generator 274 which may be configured to generate maintenance notifications to one or both of supervision system 180 and machine/user display 190.
- Regression models 240 may include one or more models configured to perform regression analyses on received machine data 210.
- ML regression trainer 230 may provide regression training data based on simulated structural responses stored in DB 236.
- Each model of the regression models 240 may be employed to perform regression analysis based on machine data 210 and simulated structural response training data to estimate damage at a plurality of locations of machine 112, including locations 114 and 116.
- Estimated damage may be output to a remaining life tracker 242 configured to monitor damage and predict a structural health of components of a frame of machine 112, implements of machine 112, and others.
- Damage aggregator 290 may receive an output from events frequency DB 270. Aggregator 290 may also be configured to receive one or more datasets stored in simulated structural responses DB 236 in order to monitor an amount of damage that accumulated over time in various locations (e.g., locations 114, 116) of machine 112. A damage validator 292 may be configured to compare aggregated damage data collected by damage aggregator 290 to historical (validation) data and thereby evaluate an accuracy of the event classification and multi-body dynamics simulation performed by event classifier models 250 and multi-body dynamics simulation 234.
- FIG. 3 illustrates a remaining life tracker 300 that may correspond to remaining life tracker 242.
- Remaining life tracker 300 may be configured to log and monitor strains, or other damage-causing conditions, at various locations of machine 112 and provide notifications to one or both of supervision system 180 and machine or user display 190.
- damage may be tracked at a plurality of locations that represent locations on an actual machine 112.
- Exemplary tracked damages may include a first cumulative damage 304, second cumulative damage 306, and third cumulative damage 308, each of which may represent a sensor-free location 116.
- a fourth exemplary damage 310 and fifth exemplary damage 312 may correspond to cumulative damages at sensor locations 114 of machine 112.
- damages at one or more locations may be tracked based on a remaining useful life (e.g., which gradually decreases over time), or based on an accumulated damage (e.g., which gradually increases over time).
- the term “damage” is understood to include either (or both) of these representations.
- a first cumulative damage 304 may correspond to 50% at a particular time, while cumulative damages 306, 308 may respectively correspond to 30% and 60%. Accordingly, each damage (e.g., damages 304-312, and others) are individually monitored and may be presented to a user of system 180 and/or display 190.
- remaining life tracker 300 may be configured to indicate a need to perform maintenance at one or more locations of machine 112, such as one or more of exemplary locations 304, 306, 308, 310, or 312.
- FIG. 3 may represent a maintenance notification displayed on system 180 or display 190 via generator 274.
- FIG. 4 illustrates an exemplary event data 400 that may correspond to events matrix 232, which may be output to multi-body dynamics simulation 234 for the acquisition of simulation data.
- Event data 400 may include data for various types of events that machine 112 may experience. Each of these types of events may have been simulated by multi-body dynamics simulation 234.
- Exemplary types of events that are stored in event data 400 may include rack events (e.g., strain introduced on frame components by encountering uneven terrain such as a pothole), bounce events, turning events, highwall (impact) events, braking events (e.g., braking force and/or deceleration greater than a predetermined threshold), and others. At least some of these events may involve one or more aspects of a work environment in which machine 112 may operate, such as a profile of the surface of terrain across which a truck may traverse (e.g., when traveling to or from a mine).
- rack events e.g., strain introduced on frame components by encountering uneven terrain such as a pothole
- bounce events e.g
- a plurality of subtypes or classes may be simulated to generate data for a particular event.
- the events and classes may be focused on a particular application, such as a mining application.
- subtypes or classes may include a pothole size (e.g., sharp or gradual), depth (e.g., 100 mm, 200 mm, etc.), a side of the machine (e.g., left or right) that encountered the pothole (see left column), and a traveling speed at which the pothole was encountered (see upper row).
- the simulation may determine strain and/or damage at various locations of machine 112, as described above.
- the occurrence rate of a particular event for an exemplary machine 112 is represented by a box in FIG. 4, with larger boxes representing increased number of occurrences of those events.
- FIG. 5 A includes a series of charts illustrating exemplary simulated data channels 500 that may be generated with multi-body dynamics simulation 144, 234, and used to characterize machine operation data.
- simulated data may be associated with a plurality of features, three of which are illustrated in FIG. 5A. These features may allow classifier models 250 to identify one or more types of events.
- simulated data channels 500 may be relevant to types of events associated with a stopping event (e.g., stopping after traveling in reverse).
- Subtypes or classes of events related to a stopping event may include striking an obstruction (e.g., an elevated wall) on the left of the machine, striking an obstruction on the right of the machine, striking an obstruction on the center or other portion of the machine, or performing braking without striking an obstruction, each of these events performed during or immediately following traveling in reverse.
- an obstruction e.g., an elevated wall
- an x-axis of charts 510, 512, and 514 may correspond to a first feature, (feature 1)
- an x-axis of charts 520, 522, and 524 may correspond to a second feature, (feature 2)
- an x-axis of charts 530, 532, and 534 may correspond to a third feature (feature 3).
- a y-axis of charts 510, 520, and 530 may correspond to feature 1
- a y-axis of charts 512, 522, and 532 may correspond to feature 2
- a y-axis of charts 514, 524, 534 may correspond to feature 3.
- Feature 1 may represent a pitch of machine 112 (e.g., by pressure sensed at struts of a suspension system), feature 2 may represent strain gauge information at one or more locations on a frame of machine 112, and feature 3 may correspond to a speed at a detected impact.
- Simulated data channels 500 may be employed to train one or more event classifier models 250 to identify a particular event, such as a stopping event.
- Characterization, or mapping, of the machine operation data to simulated data may be performed by any suitable machine learning model algorithm (implemented as one or more event classifier models 250), such as a support vector machine model, a k-nearest neighbor model, random forest machine model, or others.
- a more specific characterization or mapping may be performed by identifying a multibody dynamics simulation (e.g., signatures present in simulated data channels 500) that most closely resembles one or more event signatures in the machine operation data.
- FIG. 5B illustrates time-series data 550 corresponding to an actual operation of machine 112.
- Time-series data 550 may form an exemplary data channel of machine operation data (e.g., machine data 210).
- the horizontal axis represents time and the vertical axis represents travel speed of machine 112.
- classifier model 250 may identify events 552, 554, 556, 558, 560, and 562, and classify one or more events as a stopping-type event, and, in particular, a reverse-stopping event. By comparing one or more channels of machine operation data, which may include data 550, to simulated data, a class or subtype of the event may also be identified.
- stopping events 552, 560, and 562 may be class A events
- stopping event 558 may be a class B event
- stopping event 554 may be a class C event
- event 556 may be a class D event.
- FIGs. 6 and 7 illustrate ML assessment data 600 for validation of exemplary ML models, such as regression models 240.
- ML assessment data 600 may represent an exemplary implementation of damage validator 292.
- each circle represents a simulated event.
- Each simulated event may be associated with a particular surface characteristic (with different surface characteristics being represented by the y-axis) and speeds (with different speeds being represented by the x-axis), similar to event data 400 described above with respect to FIG. 4.
- Validation data, represented by crosses may represent a test performed on a test machine 112 equipped with sensors, including strain sensors.
- Each point of training data and each point of validation data may be associated with a respective amount of damage calculated for a particular location of machine 112, e.g., by regression models 240.
- the estimated damage calculated for one or more points (circle or cross) of ML assessment data 600 may be calculated and compared to actual damage measured by a strain gauge. The difference between these two values represents the accuracy with which events are identified and damage is estimated by system 130.
- training data 602 represented by solid circles
- validation data 612 represented by solid crosses
- Training data 604 and validation data 614 represented by circles and crosses with dashed lines, represent small amounts of error.
- the single encircled simulation of validation represents an exemplary outlier data 616.
- system 130 is configured to accurately assess damage when simulation data is employed to train one or more ML models 150 (e.g., regression models 240 and/or event classifier models 250).
- FIG. 7 includes a series of plots illustrating ML training and validation data 700 that is illustrative of the performance of ML algorithms which may be employed ML models 150.
- each plot may represent a particular location that is associated with three different channels (e.g., three sensors on a particular of location 114 configured to measure strains at this location).
- a solid line of each plot represents actual (sensed) damage or strain, while the dashed line represents an amount of strain predicted by the regression model 240 by use of the indicated ML algorithm. As can be seen in FIG.
- five exemplary ML regression algorithms useful for ML models may include MLR (multiple linear regression), random forest, as well as neural networks such as multilayer perceptron (MLP) neural networks or recurrent neural networks (RNN).
- MLP multilayer perceptron
- RNN recurrent neural networks
- neural networks employing an MLP may employ any suitable number of iterations (e.g., 200 iterations, 500 iterations, etc.).
- an NN MLP employing 500 iterations may advantageously reduce processing requirements while accurately predicting damage.
- FIG. 8 is a table illustrating exemplary work cycle data 800 that may be logged by system 130 and output by events frequency DB 270 to one or more of event tracker 280, coaching notification generator 272, or maintenance notification generator 274.
- work cycle data 800 may be logged in composite work cycles DB 282.
- Work cycle data 800 may include data representative of a plurality of types of events 810, such as dumping, loading, potholes, ditches, or turning, as illustrated in FIG. 8. Any other suitable types of events, including any events described herein (e.g., reverse-stopping events) may be included as a type of event 810.
- a composite value 820 may represent an overall frequency of the corresponding type of event 810.
- Composite value 820 may correspond to, for example, an overall number of times each respective event is expected to occur over time (e.g., a number of occurrences per hour of operation, a number of occurrences per month, a number of occurrences per miles traveled, etc.).
- Each segment 830, 832, and 834 may correspond to a simulated amount of occurrences at a respective site (e.g., a particular mine location), or a particular portion or path of the same site.
- FIG. 9 illustrates an exemplary heatmap 900 representative of a worksite (e.g., a mining worksite) on which one or more machines 112 are operated.
- Heatmap 900 may be presented on a display on supervision system 180, machine or user display 190, or both, and may correspond to an example of a notification issued by coaching notification generator 272.
- an exemplary loading location 910 may represent positions at which machines 112, such as trucks, are loaded with material.
- One or more dark-colored dots in FIG. 9 may represent an exemplary loading location.
- These machines 112 may travel along machine paths 912, such as haul roads, to one or more destination locations (light colored dots in FIG.
- work cycle monitoring system 130 may identify one or more locations 920, 922 where damage occurred by identifying events, determining which type(s) of events occurred, and estimating an amount of damage at one or more locations of machine 112 based on the identified type(s) of events.
- a geographic location at which each identified event occurred may be determined based on a location system (e.g., a GPS-based system) of machine system 110. As shown in FIG. 9, one or more locations 920 may be associated with a moderate amount of damage.
- This amount of damage may correspond to a number of damaging events at a particular location over a period of time, an average or expected damage per event at this location, or both.
- One or more locations such as encircled locations 922, may represent locations at which significant damage may occur, due to a large number of damaging events, a large amount of damage incurred per event, or both.
- FIG. 10 is a flowchart illustrating an exemplary process 1000 that may be performed for monitoring operation and/or the occurrence of damaging events in one or more machines 112.
- Step 1002 of process 1000 may include receiving machine operation data including a first data channel (e.g., machine data 210) indicative of a sensed condition at a first location of machine 112.
- This machine operation data may include or correspond to one or more work cycles of machine 112.
- machine data 210 may include one or more work cycles that include potentially damaging events.
- Machine operation data may capture the performance or condition of a component of machine 112 such as a strut, frame component, engine condition, etc., during the potentially damaging event.
- Step 1004 may include extracting some or all of the machine operation data, as described above with respect to EF extractor 220, for example.
- Data extracted from machine data 210 may include data associated with the first data channel, such as data measured by one or more sensors at a location 114.
- Step 1006 may include classifying the above-described machine operation data. Classifying may be performed by one or more event classifier models 250, as described above, to identify one or more types of events that occurred during one or more work cycles of machine 112. The one or more types of events may correspond to any of the events described herein and may involve an environment in which the machine operates.
- the environment may include a pothole, incline, decline, wall (e.g., highwall), ditch, turn, dumping action or location, loading action or location, or any other suitable environment or environmental feature.
- the environment may be a mining environment.
- Machine 112 may operate in the environment for one or more work cycles and may interact with the environment repeatedly during these work cycles as described with respect to FIG. 9, for example.
- Step 1008 may include estimating a condition at a second location such as a sensor-free location 114 of machine 112. This estimation may be performed by event classifier models 250 and/or regression models 240, and may be based on the identified type of event.
- Step 1010 may include logging the identified type of event with additional (e.g., previously identified) types of events to represent work cycles of machine 112 over time. Additionally or alternatively, the type of event may be logged via remaining life tracker 242 and/or damage validator 292. Step 1010 may include storing this information in composite work cycles DB 282, and/or providing outputs such as a coaching notifications via coaching generator 272, maintenance notifications via notification generator 274, or both. Logging performed in step 1010 may facilitate the consolidation and/or storage of data for remaining life tracker 300, work cycle data 800, heatmap 900, and/or notifications presented by supervision system 180 and/or machine or user display 190.
- FIG. 11 illustrates an implementation of a computer system 1100, which may correspond to onboard- system 120, work cycle monitoring system 130, supervision system 180, machine and/or user display 190, as well as other device(s) useful in system environment 100 or 200.
- Computer system 1100 may include a set of instructions that can be executed to cause computer system 1100 to perform any one or more of the methods or computer based functions disclosed herein.
- Computer system 1100 may operate as a standalone device or may be connected, e.g., using a network, to other computer systems or peripheral devices.
- computer system 1100 may operate in the capacity of a server or as a client user computer in a server-client user network environment, or as a peer computer system in a peer-to-peer (or distributed) network environment.
- Computer system 1100 can also be implemented as or incorporated into various devices, such as a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile device, a palmtop computer, a laptop computer, a desktop computer, a communications device, a wireless telephone, a land-line telephone, a control system, a camera, a scanner, a facsimile machine, a printer, a pager, a personal trusted device, a web appliance, a network router, switch or bridge, or any other machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine.
- PC personal computer
- PDA personal digital assistant
- computer system 1100 may be implemented using electronic devices that provide voice, video, or data communication. Further, while a single computer system 1100 is illustrated, the term “system” shall also be taken to include any collection of systems or sub-systems that individually or jointly execute a set, or multiple sets, of instructions to perform one or more computer functions.
- computer system 1100 may include a processor 1102, e.g., a central processing unit (CPU), a graphics processing unit (GPU), or both.
- processor 1102 may be a component in a variety of systems.
- processor 1102 may be part of a standard personal computer or a workstation.
- Processor 1102 may be one or more general processors, digital signal processors, application specific integrated circuits, field programmable gate arrays, servers, networks, digital circuits, analog circuits, combinations thereof, or other now known or later developed devices for analyzing and processing data.
- Processor 1102 may implement a software program, such as code generated manually (i.e., programmed).
- Computer system 1100 may include a memory 1104 that can communicate via a bus 1108.
- Memory 1104 may be a main memory, a static memory, or a dynamic memory.
- Memory 1104 may include, but is not limited to, computer readable storage media such as various types of volatile and nonvolatile storage media, including but not limited to random access memory, readonly memory, programmable read-only memory, electrically programmable readonly memory, electrically erasable read-only memory, flash memory, magnetic tape or disk, optical media and the like.
- memory 1104 includes a cache or random-access memory for processor 1102.
- memory 1104 is separate from processor 1102, such as a cache memory of a processor, the system memory, or other memory.
- Memory 1104 may be an external storage device or database for storing data.
- Memory 1104 is operable to store instructions executable by processor 1102.
- the functions, acts or tasks illustrated in the figures or described herein may be performed by the programmed processor 1102 executing the instructions stored in memory 1104.
- the functions, acts or tasks are independent of the particular type of instructions set, storage media, processor or processing strategy and may be performed by software, hardware, integrated circuits, firm-ware, micro-code and the like, operating alone or in combination.
- processing strategies may include multiprocessing, multitasking, parallel processing and the like.
- computer system 1100 may further include a display 1110, such as a liquid crystal display (LCD), an organic light emitting diode (OLED), a flat panel display, a solid-state display, a cathode ray tube (CRT), a projector, a printer or other now known or later developed display device for outputting determined information.
- Display 1110 may act as an interface for a user, to see the functioning of processor 1102, or specifically as an interface with the software stored in memory 1104 or in drive unit 1106.
- computer system 1100 may include an input device 1112 configured to allow a user to interact with any of the components of system 1100.
- Input device 1112 may be a number pad, a keyboard, or a cursor control device, such as a mouse, or a joystick, touch screen display, remote control, or any other device operative to interact with computer system 1100.
- Computer system 1100 may also or alternatively include a disk or optical drive unit 1106.
- Disk drive unit 1106 may include a computer-readable medium 1122 in which one or more sets of instructions 1124, e.g. software, can be embedded. Further, instructions 1124 may embody one or more of the methods or logic as described herein. Instructions 1124 may reside completely or partially within memory 1104 and/or within processor 1102 during execution by computer system 1100. Memory 1104 and processor 1102 also may include computer-readable media as discussed above.
- a computer-readable medium 1122 includes instructions 1124 or receives and executes instructions 1124 responsive to a propagated signal so that a device connected to a network 170 can communicate voice, video, audio, images, or any other data over the network 170. Further, the instructions 1124 may be transmitted or received over the network 170 via a communication port or interface 1120, and/or using a bus 1108.
- the communication port or interface 1120 may be a part of processor 1102 or may be a separate component. Communication port 1120 may be created in software or may be a physical connection in hardware. Communication port 1120 may be configured to connect with a network 170, external media, display 1110, or any other components in computer system 1100, or combinations thereof.
- connection with network 170 may be a physical connection, such as a wired Ethernet connection or may be established wirelessly as discussed below.
- additional connections with other components of computer system 1100 may be physical connections or may be established wirelessly.
- Network 170 may alternatively be directly connected to bus 1108.
- computer-readable medium 1122 is shown to be a single medium, the term "computer-readable medium” may include a single medium or multiple media, such as a centralized or distributed database, and/or associated caches and servers that store one or more sets of instructions.
- the term “computer-readable medium” may also include any medium that is capable of storing, encoding, or carrying a set of instructions for execution by a processor or that cause a computer system to perform any one or more of the methods or operations disclosed herein.
- Computer-readable medium 1122 is non-transitory, and may be tangible.
- Computer-readable medium 1122 can include a solid-state memory such as a memory card or other package that houses one or more nonvolatile read-only memories.
- Computer-readable medium 1122 can be a randomaccess memory or other volatile re-writable memory.
- computer-readable medium 1122 can include a magneto-optical or optical medium, such as a disk or tapes or other storage device to capture carrier wave signals such as a signal communicated over a transmission medium.
- a digital file attachment to an e-mail or other self-contained information archive or set of archives may be considered a distribution medium that is a tangible storage medium. Accordingly, the disclosure is considered to include any one or more of a computer-readable medium or a distribution medium and other equivalents and successor media, in which data or instructions may be stored.
- dedicated hardware implementations such as application specific integrated circuits, programmable logic arrays and other hardware devices, can be constructed to implement one or more of the methods described herein.
- Applications that may include the apparatus and systems of various implementations can broadly include a variety of electronic and computer systems.
- One or more implementations described herein may implement functions using two or more specific interconnected hardware modules or devices with related control and data signals that can be communicated between and through the modules, or as portions of an application-specific integrated circuit. Accordingly, the present system encompasses software, firmware, and hardware implementations.
- Network 170 may define one or more networks including wired or wireless networks.
- the wireless network may be a cellular telephone network, an 802.11, 802.16, 802.20, or WiMax network.
- networks may include a public network, such as the Internet, a private network, such as an intranet, or combinations thereof, and may utilize a variety of networking protocols now available or later developed including, but not limited to TCP/IP based networking protocols.
- Network 170 may include wide area networks (WAN), such as the Internet, local area networks (LAN), campus area networks, metropolitan area networks, a direct connection such as through a Universal Serial Bus (USB) port, or any other networks that may allow for data communication.
- WAN wide area networks
- LAN local area networks
- USB Universal Serial Bus
- the network 170 may be configured to couple one computing device to another computing device to enable communication of data between the devices.
- the network 170 may generally be enabled to employ any form of machine-readable media for communicating information from one device to another.
- Network 170 may include communication methods by which information may travel between computing devices.
- Network 170 may be divided into sub-networks. The sub-networks may allow access to all of the other components connected thereto or the sub-networks may restrict access between the components.
- Network 170 may be regarded as a public or private network connection and may include, for example, a virtual private network or an encryption or other security mechanism employed over the public Internet, or the like.
- on-board system 120 may receive sensed data from sensors at one or more sensor locations 114 and provide machine operation data, include a data channel indicative of a condition at a sensor location 114 to monitoring system 130.
- On-board system 120 may further include machine operation data generated by an ECM.
- ECM may be implemented within on-board system 120 or may be provided as a separate system in communication with on-board system 120.
- Work cycle monitoring system 130 may receive machine operation data, extract some or all of this machine operation data, and identify one or more types of events that occurred during work cycles of machine 112. This identification may be performed by use of one or more ML models, such as a plurality of ML classifier models 250 configured to identify a respective plurality of events.
- ML models may be assisted by a multi-body dynamics simulation 234 of machine 112, which may increase the accuracy and usefulness of the ML models.
- Conditions such as strain-induced damage, may be determined at one or more sensor-free locations 116 of machine 112, by use of a plurality of regression models that are provided for each of a plurality of predetermined types of events.
- the systems and methods herein allow for the use of transient data that includes information from sensors (e.g., strain sensors) and data from the ECM, which may improve event characterization.
- the use of simulation data such as a multi-body dynamics model, provides a significantly larger amount of training data for a machine learning model (such as a support vector machine), than is possible from physical testing alone.
- the information provided by the methods and systems described herein may provide feedback to users including seasonal trends, operator feedback, and locations where damage tends to occur.
- the methods described herein may be implemented by software programs executable by a computer system.
- implementations can include distributed processing, component/object distributed processing, and parallel processing.
- virtual computer system processing can be constructed to implement one or more of the methods or functionality as described herein.
- an element described herein of an apparatus embodiment is an example of a means for carrying out the function performed by the element for the purpose of carrying out the disclosure.
- any of the disclosed systems, methods, and/or graphical user interfaces may be executed by or implemented by a computing system consistent with or similar to that depicted and/or explained in this disclosure.
- aspects of the present disclosure are described in the context of computer-executable instructions, such as routines executed by a data processing device, e.g., a programmed controller or computer.
- a data processing device e.g., a programmed controller or computer.
- a data processing device e.g., a programmed controller or computer.
- Those skilled in the relevant art will appreciate that aspects of the present disclosure can be practiced with other communications, data processing, or computer system configurations, including: Internet appliances, hand-held devices, etc.
- aspects of the present disclosure may be embodied in a special purpose computer and/or data processor that is specifically programmed, configured, and/or constructed to perform one or more of the computerexecutable instructions explained in detail herein. While aspects of the present disclosure, such as certain functions, are described as being performed exclusively on a single device, the present disclosure may also be practiced in distributed environments where functions or modules are shared among disparate processing devices. Similarly, techniques presented herein as involving multiple devices may be implemented in a single device. In a distributed computing environment, program modules may be located in both local and/or remote memory storage devices.
- aspects of the present disclosure may be stored and/or distributed on non-transitory computer-readable media, including magnetically or optically readable computer discs, hard-wired or preprogrammed chips (e.g., EEPROM semiconductor chips), nanotechnology memory, biological memory, or other data storage media.
- computer implemented instructions, data structures, screen displays, and other data under aspects of the present disclosure may be distributed over the Internet and/or over other networks (including wireless networks), on a propagated signal on a propagation medium (e.g., an electromagnetic wave(s), a sound wave, etc.) over a period of time, and/or they may be provided on any analog or digital network (packet switched, circuit switched, or other scheme).
- Program aspects of the technology may be thought of as “products” or “articles of manufacture” typically in the form of executable code and/or associated data that is carried on or embodied in a type of machine- readable medium.
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Mining & Mineral Resources (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Selective Calling Equipment (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CA3190395A CA3190395A1 (en) | 2020-08-27 | 2021-07-26 | System and method for machine monitoring |
AU2021330331A AU2021330331A1 (en) | 2020-08-27 | 2021-07-26 | System and method for machine monitoring |
EP21755648.9A EP4204913A1 (en) | 2020-08-27 | 2021-07-26 | System and method for machine monitoring |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/004,546 | 2020-08-27 | ||
US17/004,546 US11656595B2 (en) | 2020-08-27 | 2020-08-27 | System and method for machine monitoring |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022046336A1 true WO2022046336A1 (en) | 2022-03-03 |
Family
ID=77358388
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2021/043108 WO2022046336A1 (en) | 2020-08-27 | 2021-07-26 | System and method for machine monitoring |
Country Status (5)
Country | Link |
---|---|
US (1) | US11656595B2 (en) |
EP (1) | EP4204913A1 (en) |
AU (1) | AU2021330331A1 (en) |
CA (1) | CA3190395A1 (en) |
WO (1) | WO2022046336A1 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7206985B2 (en) * | 2019-02-08 | 2023-01-18 | コベルコ建機株式会社 | Damage estimation device and machine learning device |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10466690B2 (en) | 2015-01-21 | 2019-11-05 | Hitachi, Ltd. | Damage estimation device |
US20200143004A1 (en) * | 2018-11-01 | 2020-05-07 | Caterpillar Inc. | Visualization of machine structure damage from machine sensor data using machine learning |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5993448B2 (en) * | 2012-04-27 | 2016-09-14 | 日立建機株式会社 | Work machine management device |
US10032117B2 (en) * | 2014-09-17 | 2018-07-24 | Caterpillar Inc. | Method for developing machine operation classifier using machine learning |
US20160098637A1 (en) * | 2014-10-03 | 2016-04-07 | Caterpillar Inc. | Automated Data Analytics for Work Machines |
US20160349152A1 (en) | 2015-06-01 | 2016-12-01 | Caterpillar Inc. | Method for aligning composite work cycle severity distributions to severity responses |
US10288530B2 (en) | 2015-06-01 | 2019-05-14 | Caterpillar Inc. | Method for improving severity estimates |
JP6453921B2 (en) * | 2017-02-03 | 2019-01-16 | ファナック株式会社 | Learning model construction device and control information optimization device |
WO2019232420A2 (en) * | 2018-06-01 | 2019-12-05 | Culvert-Iot Corporation | An intelligent tracking system and methods and systems therefor |
-
2020
- 2020-08-27 US US17/004,546 patent/US11656595B2/en active Active
-
2021
- 2021-07-26 AU AU2021330331A patent/AU2021330331A1/en active Pending
- 2021-07-26 EP EP21755648.9A patent/EP4204913A1/en active Pending
- 2021-07-26 WO PCT/US2021/043108 patent/WO2022046336A1/en unknown
- 2021-07-26 CA CA3190395A patent/CA3190395A1/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10466690B2 (en) | 2015-01-21 | 2019-11-05 | Hitachi, Ltd. | Damage estimation device |
US20200143004A1 (en) * | 2018-11-01 | 2020-05-07 | Caterpillar Inc. | Visualization of machine structure damage from machine sensor data using machine learning |
Also Published As
Publication number | Publication date |
---|---|
CA3190395A1 (en) | 2022-03-03 |
AU2021330331A1 (en) | 2023-03-30 |
US11656595B2 (en) | 2023-05-23 |
EP4204913A1 (en) | 2023-07-05 |
US20220066408A1 (en) | 2022-03-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gonzalez et al. | Slippage and immobilization detection for planetary exploration rovers via machine learning and proprioceptive sensing | |
US10288166B2 (en) | System and method for predicting and responding to soft underfoot conditions | |
US8190335B2 (en) | Performance management system for multi-machine worksite | |
AU2016201817A1 (en) | System and method for managing mixed fleet worksites using video and audio analytics | |
US11713059B2 (en) | Autonomous control of heavy equipment and vehicles using task hierarchies | |
Kassem et al. | Measuring and benchmarking the productivity of excavators in infrastructure projects: A deep neural network approach | |
Krot et al. | The identification of operational cycles in the monitoring systems of underground vehicles | |
Rashid et al. | Automated activity identification for construction equipment using motion data from articulated members | |
JP5714906B2 (en) | Systems and methods for designing haul roads | |
Akhavian et al. | Productivity analysis of construction worker activities using smartphone sensors | |
Huo et al. | Intelligent prediction for digging load of hydraulic excavators based on RBF neural network | |
US11656595B2 (en) | System and method for machine monitoring | |
Kim et al. | Hybrid kinematic–visual sensing approach for activity recognition of construction equipment | |
CN107735689A (en) | Use the system and method for accelerometer evaluation of subterranean condition of road surface | |
Mahamedi et al. | Automating excavator productivity measurement using deep learning | |
Petroutsatou et al. | Integrated prescriptive maintenance system (PREMSYS) for construction equipment based on productivity | |
GB2558255A (en) | Monitoring the operation of a work machine | |
Jakobsson et al. | Data driven modeling and estimation of accumulated damage in mining vehicles using on-board sensors | |
Väljaots et al. | Motion and energy efficiency parameters of the unmanned ground vehicle | |
Jakobsson et al. | Fatigue damage monitoring for mining vehicles using data driven models | |
JP2023541793A (en) | Track wear detection based on pressure and flow data | |
CA3193452A1 (en) | System and method for monitoring machine operations at a worksite | |
Krogerus et al. | Joint probability distributions of correlation coefficients in the diagnostics of mobile work machines | |
Kosiara et al. | Assessment of Implementation of Neural Networks in On-Board Dynamic Payload Weighing Systems | |
Shahandashti et al. | Identification of information requirements using simulation for supporting construction productivity assessment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21755648 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 3190395 Country of ref document: CA |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2021330331 Country of ref document: AU Date of ref document: 20210726 Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2021755648 Country of ref document: EP Effective date: 20230327 |